In vitro kinetics of coumarin 3,4-epoxidation: application to species differences in toxicity and carcinogenicity

Coumarin, a natural product and fragrance ingredient, is a well recognized rat liver toxicant, and dietary administration at toxic dosages increased the incidence of rat cholangiocarcinomas and parenchymal liver-cell tumors in a chronic bioassay. Hepatotoxicity in rats is site- and species-specific, and is thought to result from the formation of coumarin 3,4-epoxide and its rearrangement product, o-hydroxyphenylacetaldehyde (o-HPA). The goals of the current study were to describe the in vitro kinetics of the metabolic activation of coumarin, and determine whether species differences in susceptibility to liver injury correlate with coumarin bioactivation determined in vitro. Coumarin 3,4-epoxidation was quantified via the formation of o-HPA in pooled hepatic microsomes from female B6C3F1 mice, male F344 rats, and individual humans (n = 12 subjects), and the apparent kinetic constants for o-HPA production were calculated using nonlinear regression and fitting to either a one-enzyme or two-enzyme model. Eadie-Hofstee analyses indicated that o-HPA formation was biphasic in both rat and mouse liver. Although the apparent high affinity K:(m) in rat and mouse liver microsomes was 38.9 and 47.2 microM, respectively, the overall rate of o-HPA formation was far greater in mouse than in rat liver microsomes. Furthermore, the total clearance (CL(int)) of coumarin via o-HPA formation in mouse liver microsomes was 4-fold greater than in rat liver microsomes. Since mice are relatively resistant to hepatotoxicity, the data indicated that rates of o-HPA formation in rat and mouse liver microsomes were not directly predictive of liver toxicity in vivo, and further suggested that o-HPA detoxification played a role in modulating coumarin-mediated toxicity. The current studies also indicated that coumarin 3,4-epoxidation in human hepatic microsomes was minimal. In human liver microsomes (n = 12), the kinetics of o-HPA formation were best described by a single enzyme model, with the K(m) for o-HPA formation ranging from 1320-7420 microM. In the most active human sample, the intrinsic clearance of coumarin via the 3,4-epoxidation pathway was 1/9 and 1/38 that of the rat and mouse, respectively. The in vitro kinetics of o-HPA formation, and in particular, the large quantities of coumarin required for o-HPA production in human liver microsomes, strongly suggest that humans are unlikely to produce toxicologically relevant concentrations of this metabolite following low level coumarin exposures.